Corticosterone Regulation of 5-HT2A Receptor-Mediated Behaviors: Attenuation by Melatonin

BORIS B. GORZALKA,1 LORI A. BROTTO AND JANIE J. HONG

Department of Psychology, The University of British Columbia, Vancouver, British Columbia, Canada, V6T 1Z4

Received 15 February 1999; Accepted 26 April 1999

GORZALKA, B. B., L. A. BROTTO AND J. J. HONG. Corticosterone regulation of 5-HT2A receptor-mediated behaviors: Attenuation by melatonin. PHYSIOL BEHAV 67(3) 439–442, 1999.—The effects of chronic corticosterone treatment on sexual behavior and on wet-dog shakes (WDS), a serotonergic type 2A (5-HT2A) receptor-mediated behavior, were explored in the male rat. In addition, the effects of acute melatonin treatment, both alone and in combination with corticosterone, were investigated. Chronic injections of corticosterone resulted in an overall decrease in consummatory measures of sexual behavior, and an increase in WDS. Furthermore, although an acute injection of melatonin alone had no effect on any recorded behavior, it attenuated the effects of corticosterone on sexual behavior and WDS. The data suggest that in the context of 5-HT2A receptor-mediated behaviors, melatonin has possible implications as a 5-HT2A antagonist. © 1999 Elsevier Science Inc.

Melatonin Corticosterone Serotonin 5-HT2A receptor Sexual behavior

IN the male rat, serotonergic type 2A (5-HT2A) receptor ac- some regulatory effects on the 5-HT2A receptor. Chronic cor- tivity has been reliably found to modulate an inhibition of sex- ticosterone treatment has been demonstrated to significantly ual behavior (15). Furthermore, the frequently employed be- increase the density of central 5-HT2A receptors (9,21,26), and havior known as “wet-dog shakes” (WDS) has been found to to significantly facilitate 5-HT2A receptor-mediated behaviors be largely mediated by 5-HT2A activity, with activation of the (3,13,18). 5-HT2A receptor producing a robust facilitation of WDS in the The hormone, melatonin, has also been explored in the rat (29). WDS, which consist of a rotational shudder of the up- context of 5-HT2A receptor regulation. Eison et al. (8) demon- per body (2), resemble the purposeful movement, as seen in strated a dose-dependent attenuation by melatonin of the in- dogs, and have reliably been used as a behavioral assay of crease in WDS induced by a 5-HT2A receptor agonist. More- 5-HT2A receptor activity (29). A strong, inverse correlation over, radioligand binding techniques demonstrated that melatonin exists between male sexual behavior and WDS, such that acti- reduced the concentration-dependent 5-HT2A-mediated phos- vation of the 5-HT2A receptor produces an increase in WDS, phoinositide (PI) hydrolysis response to 5-HT2A agonists and a concomitant decrease in sexual behavior (27). As a re- without altering central 5-HT2A receptor density (8). Further- sult, spontaneously occurring WDS have been frequently em- more, investigations that demonstrate a similarity in the be- ployed as a noninvasive index of 5-HT2A activity during sexual havioral effects of melatonin to 5-HT2A antagonists (10) are behavior. consistent with melatonin’s putative 5-HT2A antagonism. Unique to the 5-HT2A receptor is its “nonclassical” mecha- Given the apparently opposing effects of corticosterone nism of regulation (24). The receptor’s relative resistance to and melatonin on 5-HT2A receptor activity, it seems reason- upregulation in response to serotonergic manipulation has able to predict that concurrent administration of melatonin triggered an interest in nonserotonergic methods of its regula- would attenuate the behavioral effects of corticosterone treat- tion. For example, the adrenal hormone, corticosterone, has ment. Furthermore, because drugs that antagonize 5-HT2A re- been explored in this context, and has been found to exert ceptor activity, including ritanserin, pirenperone, ketanserin,

1To whom requests for reprints should be addressed. E-mail: [email protected] 439

440 GORZALKA, BROTTO AND HONG and nefazodone, exert no effects on sexual behavior (28) and val. In addition, the frequency of WDS was tallied for a 30- WDS (18), it is expected that neither should melatonin, a pu- min observation period. tative 5-HT2A antagonist, exert effects on these measures All testing was conducted during the middle third of the when administered alone. The aim of the present study is to dark cycle by trained observers blind to the experimental con- investigate the effects of corticosterone, both alone and in ditions of the animals. A priori predictions that 1) melatonin combination with melatonin, on sexual behavior and WDS, should attenuate the effects of corticosterone, 2) melatonin and to explore the possibility of 5-HT2A antagonism as a result alone should not exert any behavioral effects, and 3) cortico- of melatonin treatment. sterone alone should produce behavioral effects that differ from the other three conditions, and that are similar to each other, permit use of the statistical method of nondirectional Planned MATERIALS AND METHODS Contrasts (11), with a significance level preset at p Ͻ 0.05. Subjects RESULTS Eighty Long–Evans male rats (Charles River Canada Inc., Quebec), were obtained at 5 weeks of age. Prior to testing, Table 1 suggests that corticosterone inhibited sexual be- males were screened for copulatory proficiency, and those havior and increased WDS, and that these effects were atten- displaying consistently vigorous sexual activity were selected uated by melatonin. Planned comparisons revealed that corti- for the study. The screening procedure resulted in 68 males costerone significantly decreased the frequency of ejaculations being employed in the study. At the time of testing, males compared to the control, melatonin-treated, and corticoste- were 4.5 months of age and 475 g on average. In addition, 18 rone combined with melatonin-treated groups, as a whole, sexually experienced female rats were used to elicit sexual be- t(64) ϭ 2.03, p ϭ 0.047. Also, melatonin significantly attenu- havior in the males. Females were previously bilaterally ova- ated the effects of corticosterone on ejaculations, t(64) ϭ 2.17, riectomized at 3 months of age using standard surgical proce- p ϭ 0.034. Melatonin alone did not affect ejaculation fre- dures while anesthetized with ketamine HCl (75 mg/kg) and quency, p Ͼ 0.05. xylazine (7 mg/kg) obtained from the UBC Animal Care Cen- A similar pattern was observed for ejaculation latency. tre, Vancouver, Canada. Melatonin significantly blocked the effects of corticosterone, All rats were housed in same-sex groups of three or four, t(64) ϭ Ϫ2.24, p ϭ 0.028. Although not statistically signifi- in standard wire mesh cages, and were allowed free access to cant, corticosterone increased the ejaculation latency com- Purina Rat Chow and water. Colony conditions were main- pared to the other three conditions as a group, t(64) ϭ tained at 21 Ϯ 1ЊC, and animals were kept on a reverse 12/12- Ϫ1.848, p ϭ 0.069. Again, melatonin alone had no effect on h light cycle (lights off at 0900 h). ejaculation latency, p Ͼ 0.05. For mounting behavior, corticosterone significantly in- Injection Procedure creased the frequency of mounts compared to the other three conditions as a group, t(33) ϭ Ϫ2.503, p ϭ 0.017. Whereas Corticosterone-21-acetate (Sigma Chemical Co., Chicago, melatonin appeared to block the effects of corticosterone on IL) was suspended in propylene glycol (20 mg/mL) and mela- mounts, this effect did not quite reach statistical significance, tonin (Sigma Chemical Co) was dissolved in a solution of 20% p Ͼ 0.05. Although similar trends were observed on some dimethyl sulfoxide (DMSO) and saline (6 mg/mL). Cortico- other measures of sexual behavior, there was no significant ef- sterone or the vehicle, propylene glycol, was injected subcuta- fect of either corticosterone or melatonin treatment on mount neously for 14 days (1 mL/kg). On the 15th day, animals re- latency, intromission latency, intromission frequency, and post- ceived an intraperitoneal injection of either melatonin (1 mL/ ejaculatory interval. kg), or the vehicle, DMSO, 45 min prior to testing. In addi- Corticosterone significantly increased WDS relative to the tion, given previous evidence that the frequency of spontane- other three conditions as a group, t(64) ϭ Ϫ3.969, p ϭ 0.001. ously occurring WDS is quite low in males engaging in copula- In addition, melatonin significantly attenuated the effects of tory behavior (4), all animals received as injection (1 mL/kg) corticosterone on WDS, t(64) ϭ Ϫ3.803, p ϭ 0.001. Again, of the 5-HT receptor agonist, (ϩ)1-(2,5 dimethyl-4-iodo- 2A melatonin alone had no effect on WDS, p Ͼ 0.05. phenyl)-2-aminopropane (DOI; Research Biochemicals In- ternational), to amplify this effect. DOI was dissolved in 0.9% saline (1.25 mg/mL), 30 min prior to testing. Male subjects were randomly assigned to one of four treatment groups: 1) DISCUSSION propylene glycol and DMSO, n ϭ 18; 2) propylene glycol and The present results support previous findings of a cortico- melatonin, n ϭ 20; 3) corticosterone and DMSO, n ϭ 15; and sterone-induced inhibition of sexual behavior in the male rat 4) corticosterone and melatonin, n ϭ 15. (13) and facilitation of WDS (3,18). The chronic corticosterone regimen employed in these studies results in plasma cor- Behavioral Testing Procedure ticosterone levels that are similar to those produced after a Females were injected subcutaneously with 10 ␮g estradiol chronic stressor (22), and various stressors have been found to benzoate (Sigma Chemical Co.) 2 days before testing, and 500 inhibit sexual behavior and facilitate WDS in the male rat ␮g progesterone (Sigma Chemical Co.), 4 h before testing. (4,14). Although acute melatonin treatment alone exerted no Hormones were dissolved in 0.1 mL peanut oil. Males were effect on male sexual activity, it completely reversed the corti- tested in Plexiglas chambers (30 ϫ 30 ϫ 45 cm in height) cov- costerone-induced inhibition of ejaculatory behavior. These ered with contact bedding. are the first reported data to suggest that melatonin may pro- Males were given 5 min to habituate to the chambers be- tect against the debilitating effects of chronic stress on male fore being presented with a receptive female. Measures of sexual behavior. sexual behavior included: mount, intromission, and ejacula- The observed behavioral responses to chronic corticoste- tion frequencies and latencies, and the postejaculatory inter- rone treatment are consistent with a 5-HT2A receptor-medi-

CORTICOSTERONE, MELATONIN, AND 5-HT2A RECEPTORS 441

TABLE 1 EFFECTS OF CHRONIC CORTICOSTERONE TREATMENT AND ACUTE MELATONIN TREATMENT ON FREQUENCIES OF MOUNTS, INTROMISSIONS, AND EJACULATIONS, LATENCIES OF MOUNTS (ML), INTROMISSIONS (IL), AND EJACULATIONS (EL) IN SECONDS, POSTEJACULATORY INTERVAL (PEI) IN SECONDS, AND FREQUENCY OF WET DOG SHAKES (WDS) IN MALE RATS

No Corticosterone Corticosterone

No Melatonin Melatonin No Melatonin Melatonin

Mounts* 3.25 Ϯ 0.75 4.58 Ϯ 1.06 8.33 Ϯ 1.73 5.55 Ϯ 1.07 Intromissions 9.00 Ϯ 1.20 9.33 Ϯ .83 10.67 Ϯ .92 8.27 Ϯ .82 Ejaculations* 1.17 Ϯ 0.33 1.50 Ϯ 0.34 0.67 Ϯ 0.23 1.73 Ϯ 0.37 ML 914.4 Ϯ 187.0 877.9 Ϯ 178.8 1108.0 Ϯ 188.5 601.1 Ϯ 199.6 IL 970.4 Ϯ 195.9 896.2 Ϯ 176.9 1165.7 Ϯ 193.2 724.6 Ϯ 197.2 EL† 1077.6 Ϯ 196.1 854.5 Ϯ 179.3 1276.7 Ϯ 176.5 641.7 Ϯ 194.7 PEI 1181.3 Ϯ 168.1 974.4 Ϯ 166.9 1301.5 Ϯ 166.6 853.1 Ϯ 194.4 WDS* 13.1 Ϯ 2.1 13.4 Ϯ 1.7 25.7 Ϯ 5.0 9.2 Ϯ 2.5

Values represent means Ϯ SEMs. *Denotes that the group tested with corticosterone alone differed significantly from the other three groups, p Ͻ 0.05. †Denotes that the group treated with corticosterone alone differed from the group treated with both corticosterone and melatonin, p Ͻ 0.05.

ated mechanism suppressing male sexual behavior while in- receptor upregulation independent of changes in 5-HT levels creasing WDS (27). Both chronic corticosterone treatment (21). The previous suggestion that hormones of the HPA axis (9,21,26) and corticosterone elevation induced by a chronic may play a major role in the regulation of 5-HT2A receptor stressor (22) significantly increase the density of central 5-HT2A density (4) may also be extended to include melatonin (8). receptors. WDS provide an index of 5-HT2A receptor activity A melatonin–corticosterone interaction has been widely (29); therefore, the present observation of increased WDS af- investigated, and melatonin has been suggested to exert anti- ter corticosterone administration supports the notion of a cor- glucocorticoid effects (1). Melatonin has been implicated as ticosterone–5-HT2A receptor interaction. protecting against the detrimental effects of elevated gluco- Melatonin significantly attenuated the effects of corticos- corticoids on the hypothalamic–pituitary–adrenal axis (19,20). terone treatment on measures of sexual behavior and WDS, In addition, Chuang and Lin (6) have reported that the ther- albeit in the presence of the 5-HT2A receptor agonist, DOI. mal stress-induced increases in locomotor activity are effec- This observation further supports previous observations that tively attenuated by acute melatonin treatment. However, implicate melatonin as a 5-HT2A antagonist. Melatonin has melatonin has no effect on corticosterone-induced thymus re- previously been shown to attenuate agonist-induced WDS gression (16). Therefore, it remains possible that antigluco- (8), and radioligand investigations revealed that it blocks the corticoid effects of melatonin may not become apparent be- 5-HT2A second messenger system. In addition, administration haviorally until pharmacological doses are administered, and of the nonselective 5-HT2A antagonist methysergide has been that at physiological levels, there may be little or no coupling reported to induce behavioral changes similar to those found between these hormones. These antiglucocorticoid effects of after treatment with melatonin (10). Opposite effects between melatonin may be mediated, in part, by a 5-HT2A receptor an- melatonin and 5-HT2A antagonists have also been discovered. tagonism. For example, pretreatment with the antidepressant, mianserin, Elevated circulating glucocorticoids are one physiological a 5-HT2A antagonist, prevents the effects of intraaccumbens response to stress in humans, and stress has been identified as melatonin administration on locomotor activity and sniffing one of the predisposing factors for major depression (12). behavior (7). However, this effect of mianserin may be attrib- Postmortem studies on the brains of depressed suicide victims utable to activity at 5-HT2 receptors other than the 5-HT2A have found an increase in 5-HT2A receptor density (23), and subtype, and does not necessarily generalize to sexual behav- the therapeutic onset of most antidepressants occurs with the ior. Therefore, it remains plausible that melatonin exerts concurrent downregulation of the 5-HT2A receptor (25). Me- some effects via a 5-HT2A receptor-mediated mechanism. latonin has been implicated as a mood enhancer, and as pos- Although there have been extensive investigations as to sessing some possible therapeutic benefits to individuals with the functional interaction between melatonin and serotonin, depression (17). As a preliminary hypothesis, melatonin may corticosterone has been reported to have little or no effect on play a role in mood elevation by acting as an antagonist at the 5-HT metabolism (5). Therefore, the increase in WDS and de- 5-HT2A receptor. crease in male rat sexual behavior following corticosterone administration are likely due to a specific receptor mechanism rather than an effect on serotonin, per se. Because these effects were blocked to varying degrees by melatonin, this fur- ACKNOWLEDGEMENTS ther supports an action on the 5-HT2A receptor, and not on This work was supported by a grant to B. Gorzalka from the Nat- levels of serotonin. Moreover, radioligand studies have dem- ural Sciences and Engineering Research Council of Canada onstrated that chronic corticosterone treatment induces 5-HT2A (NSERC) and a NSERC Predoctoral Fellowship to L. Brotto. 442 GORZALKA, BROTTO AND HONG

REFERENCES 1. Aoyama, H.; Mori, W.; Mori, N.: Anti-glucocorticoid effects of kind, D.; Sengos, G.; Zhou, D.; Huether, G.: No evidence for a melatonin in young rats. Acta Pathol. Jpn. 36:423–428; 1986. physiological coupling between melatonin and glucocorticoids. 2. Bedard, P.; Pycock, C. J.: Wet-dog shake behaviour in the rat: A Psychopharmacology (Berlin) 133:313–322; 1997. possible quantitative model of central 5-hydroxytryptamine activ- 17. Halbreich, U.: Hormonal interventions with psychopharmacolog- ity. Neuropharmacology 16:663–670; 1977. ical potential: An overview. Psychopharmacol. Bull. 33:281–286; 3. Berendsen, H. H. G.; Kester, R. C. H.; Peeters, B. W. M. M.; 1997. Broekkamp, C. L. E.: Modulation of 5-HT receptor subtype- 18. Hanson, L. A.; Gorzalka, B. B.; Brotto, L. A.: The antidepres- mediated behaviours by corticosterone. Eur. J. Pharmacol. sant, nefazodone, attenuates corticosterone-induced increases in 308:103–111; 1996. 5-HT2A mediated behaviors in the female rat. Eur. J. Pharmacol. 4. Brotto, L. A.; Gorzalka, B. B.; Hanson, L. A.: Effects of housing 342:163–165; 1998. conditions and 5-HT2A activation on male rat sexual behavior. 19. Konakchieva, R.; Mitev, U.; Almeida, O. F. X.; Patchev, V. K.: Physiol. Behav. 63:475–479; 1998. Chronic melatonin treatment and the hypothalamo–pituitary– 5. Chaouloff, F.: Physiopharmacological interactions between stress adrenal axis in the rat. Attenuation of the secretory response to hormones and central serotonergic systems. Brain Res. Rev. stress and effects on hypothalamic neuropeptide content and 18:1–32; 1993. release. Biol. Cell. 89:587–596; 1997. 6. Chuang, J. I.; Lin, M. T.: Pharmacological effects of melatonin 20. Konakchieva, R.; Mitev, U.; Almeida, O. F. X.; Patchev, V. K.: treatment on both locomotor activity and brain serotonin release Chronic melatonin treatment counteracts glucocorticoid-induced in rats. J. Pineal Res. 17:11–16; 1994. dysregulation of the hypothalamic–pituitary–adrenal axis in the 7. Durlach–Misteli, C., Van Ree, J. M.: Dopamine and melatonin in rat. Neuroendocrinology 67:171–180; 1998. the nucleus accumbens may be implicated in the mode of action 21. Kuroda, Y.; Mikuni, M.; Ogawa, T.; Takahashi, K.: Effect of of antidepressant drugs. Eur. J. Pharmacol. 217:15–21; 1992. ACTH, adrenalectomy and the combination treatment on the 8. Eison, A. S.; Freeman, R. P.; Guss, V. B.; Mullins, U. L.; Wright, density of 5-HT2 receptor binding sites in the neocortex of rat R. N.: Melatonin agonists modulate 5-HT2A receptor-mediated forebrain and 5-HT2 receptor-mediated wet-dog shake behaviors. neurotransmission: Behavioral and biochemical studies in the rat. Psychopharmacology (Berlin) 108:27–32; 1992. J. Pharmacol. Exp. Ther. 273:304–308; 1995. 22. McKittrick, C. R.; Blanchard, D. C.; Blanchard, R. J.; McEwen, 9. Fernandes, C.; McKittrick, C. R.; File, S. E.; McEwen, B. S.: B. S.; Sakai, R. R.: Serotonin receptor binding in a colony model Decreased 5-HT1A and increased 5-HT2A receptor binding after of chronic social stress. Biol. Psychiatry 37:383–393; 1995. chronic corticosterone associated with a behavioural indication of 23. Pandey, G. N.; Pandey, S. C.; Dwivedi, Y.; Sharma, R. P.; Janicak, depression but not anxiety. Psychoneuroendocrinology 22:477– P. G.; Davis, J. M.: Platelet serotonin-2A receptors: A potential 491; 1997. biological marker for suicidal behavior. Am. J. Psychiatry 152:850– 10. Gaffori, O.; Van Ree, J. M.: Serotonin and antidepressant drugs 855; 1995. antagonize melatonin-induced behavioural changes after injec- 24. Sanders-Bush, E.: Adaptive regulation of central serotonin tion into the nucleus accumbens of rats. Neuropharmacology receptors linked to phosphoinositide hydrolysis. Neuropsycho- 24:237–244; 1985. pharmacology 3:411–416; 1990. 11. Glass, G. V.; Hopkins, K. D.: Statistical methods in education and 25. Sugrue, M. F.: Do antidepressants possess a common mechanism psychology, 3rd ed. Needham Heights: Allyn and Bacon; 1996. of action? Biochem. Pharmacol. 32:1811–1817; 1983. 12. Gold, P. W.; Goodwin, F. K.; Chrousos, G. P.: Clinical and bio- 26. Takao, K.; Nagatani, T.; Kitamura, Y.; Yamawaki, S.: Effects of chemical manifestations of depression: Relation to the neurobiol- corticosterone on 5-HT1A and 5-HT2 receptor binding and on the ogy of stress. N. Engl. J. Med. 319:413–420; 1988. receptor-mediated behavioral responses of rats. Eur. J. Pharma- 13. Gorzalka, B. B.; Hanson, L. A.: Sexual behavior and wet dog col. 333:123–128; 1997. shakes in the male rat: Regulation by corticosterone. Behav. 27. Watson, N. V.; Gorzalka, B. B.: Relation of spontaneous wet dog Brain. Res. 97:143–151; 1998. shakes and copulatory behavior in male rats. Pharmacol. Bio- 14. Gorzalka, B. B.; Hanson, L. A.; Brotto, L. A.: Chronic stress chem. Behav. 37:825–829; 1990. effects on sexual behavior in male and female rats: Mediation by 28. Watson, N. V.; Gorzalka, B. B.: DOI-induced inhibition of copu- 5-HT2A receptors. Pharmacol. Biochem. Behav. 61:405–412; 1998. latory behavior in male rats: Reversal by 5-HT2A antagonists. 15. Gorzalka, B. B.; Mendelson, S. D.; Watson, N. V.: Serotonin Pharmacol. Biochem. Behav. 39:605–612; 1991. receptor subtypes and sexual behavior. Ann. NY Acad. Sci. 29. Yap, C. Y.; Taylor, D. A.: Involvement of 5-HT2 receptors in the 600:435–446; 1990. wet-dog shake behaviour induced by 5-hydroxytryptophan in the 16. Hajak, G.; Rodenbeck, A.; Ehrenthal, H.; Leonard, S.; Wede- rat. Neuropharmacology 22:801–804; 1983.